RNAsum: a tool for personalised genome and transcriptome interpretation for improved cancer diagnostics

Abstract

BackgroundThe integration of whole-genome sequencing (WGS) and whole-transcriptome sequencing (WTS) has revolutionized cancer diagnostics by enabling comprehensive molecular profiling of tumours. WGS uncovers genomic alterations such as single nucleotide variants, structural variants, and copy number changes, whereas WTS reveals their functional consequences through expression profiles and fusion detection. Together, these technologies offer unparalleled potential to guide precision oncology by identifying actionable biomarkers and stratifying patients for targeted therapies or clinical trials. Recent studies have shown that nearly half of patients experience improved clinical outcomes when treatment is guided by combined WGTS analysis. However, dedicated tools for automated single-patient integration and reporting of WGTS data within routine clinical bioinformatics workflows remain limited, representing a significant gap in personalised cancer sample interpretation. ResultsWe developed RNAsum, an open-source tool for integrating and interpreting whole-genome sequencing (WGS) and whole-transcriptome sequencing (WTS) data from individual cancer patient samples. RNAsum compares patient data to The Cancer Genome Atlas (TCGA) cohorts, integrating quantitative expression data with genomic findings to corroborate and prioritise clinically relevant alterations and enhance diagnostic accuracy. Clinical applicability evaluation performed across 60 patients demonstrated that 68% (140/205) of the clinically reportable variants identified by WGS, spanning copy number changes, truncating mutations and gene fusions, were supported at the RNA level as detected by WTS and reported by RNAsum. Case studies further highlight the ability of RNAsum to support clinically reportable variants, refine diagnoses, and identify novel therapeutic targets, particularly in complex cases involving multiple genomic alterations and drug resistance mechanisms. ConclusionRNAsum effectively bridges the gap between genome and transcriptome analyses, significantly advancing the integration of multiomics data in personalized cancer care. Its ability to corroborate clinically reportable variants at the RNA level and elucidate complex alterations, including those driving drug resistance, highlights RNAsums potential to improve molecular tumour profiling and support clinical decision-making. Freely available as an R package on GitHub at https://github.com/umccr/RNAsum, RNAsum provides an accessible and scalable solution for researchers and clinicians, representing a significant step towards the routine application of integrated genomic and transcriptomic analyses in precision oncology.